{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:MSZHMKLBBXQUCRYVQXOYB6ENBE","short_pith_number":"pith:MSZHMKLB","canonical_record":{"source":{"id":"2406.17626","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T15:13:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"57a1ba65fbdb7147e9f1cfdeec50e94dbb7bfcf3a051bccb533b3d545a39244a","abstract_canon_sha256":"f2babaa46f051ac9322fab81e39c051421b18bb4bb4c7d19d61c2ff0fd609235"},"schema_version":"1.0"},"canonical_sha256":"64b27629610de141471585dd80f88d0916607f459f99886206a73a58695b1fbe","source":{"kind":"arxiv","id":"2406.17626","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.17626","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"arxiv_version","alias_value":"2406.17626v1","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17626","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"pith_short_12","alias_value":"MSZHMKLBBXQU","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"pith_short_16","alias_value":"MSZHMKLBBXQUCRYV","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"pith_short_8","alias_value":"MSZHMKLB","created_at":"2026-07-05T08:36:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:MSZHMKLBBXQUCRYVQXOYB6ENBE","target":"record","payload":{"canonical_record":{"source":{"id":"2406.17626","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T15:13:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"57a1ba65fbdb7147e9f1cfdeec50e94dbb7bfcf3a051bccb533b3d545a39244a","abstract_canon_sha256":"f2babaa46f051ac9322fab81e39c051421b18bb4bb4c7d19d61c2ff0fd609235"},"schema_version":"1.0"},"canonical_sha256":"64b27629610de141471585dd80f88d0916607f459f99886206a73a58695b1fbe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:37.993204Z","signature_b64":"AGoE/euLEJIy5i6RyskLMLfxmMrPBkI28yEKOfv7w02W31s4iZRQCsXMrABM+R7VfXf2gOha6e7UxTXf3ITsDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64b27629610de141471585dd80f88d0916607f459f99886206a73a58695b1fbe","last_reissued_at":"2026-07-05T08:36:37.992761Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:37.992761Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.17626","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:36:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yNGBVWTORHxRUYNNSYtw04ruxHazMp/EUwL26ci5s2fWU6vuiKl9ryihgbWxHjw+olWLhG2JycWRHykTYS2oBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:56:04.252216Z"},"content_sha256":"a87d71d2f49cc2d58a006800d2f15a155e55a4800047938c18d56bbf09c6bc55","schema_version":"1.0","event_id":"sha256:a87d71d2f49cc2d58a006800d2f15a155e55a4800047938c18d56bbf09c6bc55"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:MSZHMKLBBXQUCRYVQXOYB6ENBE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CoSafe: Evaluating Large Language Model Safety in Multi-Turn Dialogue Coreference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Erxin Yu, Fei Mi, Jing Li, Lanqing Hong, Ming Liao, Siqi Wang, Zuchen Gao","submitted_at":"2024-06-25T15:13:02Z","abstract_excerpt":"As large language models (LLMs) constantly evolve, ensuring their safety remains a critical research problem. Previous red-teaming approaches for LLM safety have primarily focused on single prompt attacks or goal hijacking. To the best of our knowledge, we are the first to study LLM safety in multi-turn dialogue coreference. We created a dataset of 1,400 questions across 14 categories, each featuring multi-turn coreference safety attacks. We then conducted detailed evaluations on five widely used open-source LLMs. The results indicated that under multi-turn coreference safety attacks, the high"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17626","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2406.17626/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:36:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KXKKkxT6hLDBgIJNMV8OQmRLHeZZ1vvgM3vznz96z0WnjpHLkKHfo+4E8E2tRdzjwSqFpjF7AEQQ+SmeekY1BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:56:04.252709Z"},"content_sha256":"f6f24ae1a235c959e35f41f6464f0397bb7fff2604d7a7b5dd0451581b7e85c3","schema_version":"1.0","event_id":"sha256:f6f24ae1a235c959e35f41f6464f0397bb7fff2604d7a7b5dd0451581b7e85c3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE/bundle.json","state_url":"https://pith.science/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T06:56:04Z","links":{"resolver":"https://pith.science/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE","bundle":"https://pith.science/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE/bundle.json","state":"https://pith.science/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MSZHMKLBBXQUCRYVQXOYB6ENBE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:MSZHMKLBBXQUCRYVQXOYB6ENBE","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"f2babaa46f051ac9322fab81e39c051421b18bb4bb4c7d19d61c2ff0fd609235","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T15:13:02Z","title_canon_sha256":"57a1ba65fbdb7147e9f1cfdeec50e94dbb7bfcf3a051bccb533b3d545a39244a"},"schema_version":"1.0","source":{"id":"2406.17626","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.17626","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"arxiv_version","alias_value":"2406.17626v1","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17626","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"pith_short_12","alias_value":"MSZHMKLBBXQU","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"pith_short_16","alias_value":"MSZHMKLBBXQUCRYV","created_at":"2026-07-05T08:36:37Z"},{"alias_kind":"pith_short_8","alias_value":"MSZHMKLB","created_at":"2026-07-05T08:36:37Z"}],"graph_snapshots":[{"event_id":"sha256:f6f24ae1a235c959e35f41f6464f0397bb7fff2604d7a7b5dd0451581b7e85c3","target":"graph","created_at":"2026-07-05T08:36:37Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2406.17626/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) constantly evolve, ensuring their safety remains a critical research problem. Previous red-teaming approaches for LLM safety have primarily focused on single prompt attacks or goal hijacking. To the best of our knowledge, we are the first to study LLM safety in multi-turn dialogue coreference. We created a dataset of 1,400 questions across 14 categories, each featuring multi-turn coreference safety attacks. We then conducted detailed evaluations on five widely used open-source LLMs. The results indicated that under multi-turn coreference safety attacks, the high","authors_text":"Erxin Yu, Fei Mi, Jing Li, Lanqing Hong, Ming Liao, Siqi Wang, Zuchen Gao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T15:13:02Z","title":"CoSafe: Evaluating Large Language Model Safety in Multi-Turn Dialogue Coreference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17626","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a87d71d2f49cc2d58a006800d2f15a155e55a4800047938c18d56bbf09c6bc55","target":"record","created_at":"2026-07-05T08:36:37Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"f2babaa46f051ac9322fab81e39c051421b18bb4bb4c7d19d61c2ff0fd609235","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T15:13:02Z","title_canon_sha256":"57a1ba65fbdb7147e9f1cfdeec50e94dbb7bfcf3a051bccb533b3d545a39244a"},"schema_version":"1.0","source":{"id":"2406.17626","kind":"arxiv","version":1}},"canonical_sha256":"64b27629610de141471585dd80f88d0916607f459f99886206a73a58695b1fbe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"64b27629610de141471585dd80f88d0916607f459f99886206a73a58695b1fbe","first_computed_at":"2026-07-05T08:36:37.992761Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:36:37.992761Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AGoE/euLEJIy5i6RyskLMLfxmMrPBkI28yEKOfv7w02W31s4iZRQCsXMrABM+R7VfXf2gOha6e7UxTXf3ITsDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:36:37.993204Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.17626","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a87d71d2f49cc2d58a006800d2f15a155e55a4800047938c18d56bbf09c6bc55","sha256:f6f24ae1a235c959e35f41f6464f0397bb7fff2604d7a7b5dd0451581b7e85c3"],"state_sha256":"7d2a400c69b5ce37f33f1dbb22892129c2a34ecee30de45c645dfa7fc466457a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cXmE1cov+UTkHPsOqNWZMhkhKxklW1cPnAza/udADYsA7IwT4ke6ylri8y0jRuyiQ1GCAFaxKCZbB33YGVtwAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T06:56:04.255856Z","bundle_sha256":"2cca2fe5bdd6462e355a7f9749a044accde472efc5fdb25878194a17119b19b2"}}